The enzyme ornithine decarboxylase (ODC) catalyses the first step in mammalian polyamine biosynthesis and appears to play a central role in tumor promotion. Inhibition of ODC induction in animal models also blocks tumor formation in several organs. Difluoromethylornithine (DFMO) is a suicide inhibitor of ODC with demonstrated ability to inhibit induction of ODC in human skin biopsies and reduce polyamine levels in rectal mucosae at doses a fraction of the maximally tolerated doses. At a single daily dose of 0.5 g/m2 DFMO, the current investigators have treated 9 subjects for up to 12 months with no significant toxicity and suppression of phorbol-ester induced ODC in skin to less than 50% of baseline levels. In the proposed research, the goal is to determine whether this dose of DFMO significantly suppresses polyamine and ODC activity levels in colorectal and prostate tissues. 40 subjects at risk for colorectal cancer will be evaluated in a 1 year double blind placebo-controlled trial, during which proximal and distal colorectal mucosal specimens will be evaluated before and after treatment for ODC and polyamine levels. 25 subjects with early stage prostate or bladder cancers will be evaluated in a placebo controlled, double blind trial with measurement of polyamine and ODC levels in needle core and larger specimens of malignant and nonmalignant prostate tissue at the time of definitive surgery, after 14 days of DFMO treatment. Finally, 46 subjects with minimally elevated prostate specific antigen (PSA) levels will be studied in a placebo controlled trial to evaluate ODC and polyamine levels in needle core biopsies of prostate tissues after 14 days of treatment. The approximately 40 patients of this group, who are not found to have malignancy, will all be treated with DFMO, and PSA, prostatic acid phosphate, transrectal prostatic ultrasound, PSA density and plasma testosterone will be evaluated over 1 year. At the conclusion of this research the investigators will have specific target tissue data for the colon and prostate for an apparently non toxic dose of DFMO. In addition, further toxicity data will be obtained from a total of 50-60 individuals treated on this 0.5 g/m2/day single dose schedule for one year in this study, which should make clear the likely safety of this drug for large chemoprevention studies. Finally, with the detailed data on tissue specific biological markers obtained in this study, the creation of efficient designs for phase III chemoprevention trials with DFMO will be more feasible.